A storage system typically includes a plurality of storage disks, and these storage disks can be organized in a disk array. For example, a redundant array of independent disks (RAID) is a data storage virtualization storage technology that combines multiple storage disks into a single logic unit for the purpose of data redundancy backup and/or performance improvement. Taking a typical RAID 5 as an example, one disk array may correspond to a group of storage disks, and the RAID 5 uses the space of the group of storage disks for storing data and verifying information. In addition, there is often a spare disk for the disk array. Considering the consistency of input/output (I/O), the type of the spare disk is usually the same with the type of the storage disk in the disk array (for example, a disk). When a certain storage disk in the disk array fails, the spare disk can be used to temporarily replace the failed disk. In this case, the data on the failed disk will be recovered onto the spare disk for reconstructing the disk array.
In that case that the failed disk is a large-capacity disk, it takes much time to perform the reconstruction procedure for the disk array using a spare disk. Besides, when the storage disk fails, the disk array has a degraded operation performance. The above cases bring adverse effects to data safety and consistency of the disk array.